RFM stands for “recency, frequency, and monetary” as a term that can be used in reference to recent customer transactions when analyzed for their value. This type of purchase analysis can be used to rank and group customers based on these variables to identify the best customers for similar, future purchases and perform targeted marketing campaigns accordingly. 

RFM analysis ranks each customer on the following factors and assigns a score, commonly on a scale of 1-5:

  • Recency analysis – Typically this metric is evaluated in days, and the more recent the higher ranked, but depending on the product or service in question this could be adjusted, e.g. if a typical renewal has an annual lifecycle then 12-months could be considered a higher-value recency period.
  • Frequency analysis – A count of purchases for each customer. The more purchases, the higher-ranked, since customers that have purchased at least once are typically more likely to purchase again.
  • Monetary analysis – A count of the sum of purchases for each customer. The more purchased, the higher-ranked, since customers that have purchased with a high spend are typically more likely to purchase again and are considered high-value customers.

To assign a score based on identified ranges there are multiple options available including leveraging a Business Intelligence tool or applying a more advanced ETL process that could extract, transform and load the data back in as the translated score into either an extension tables or a custom object, depending on the MAP being used. In a MAP like Eloqua, this can be done programmatically from within the platform by evaluating contacts against the ranges using filter criteria and applying a score via update rule into a separate field. 

Approaches vary in how to rank customers based on this information but include simply averaging the sum value of all three variables or providing some type of weighting to each score out of the total. In either scenario, the resulting score can then be used to rank customers for a simplistic predictive analysis of future purchases they’ll be likely to make. 

Based on these scores, commonly evaluated customer segments like these can be built:

  • New Customers will have high recency but low frequency scores. Targeted follow-up campaigns can increase the chance of converting them into repeat customers.
  • Core Customers will have high recency, frequency, and monetary scores. Instead of discounts, marketing efforts focused on a value-add will perform best, e.g. cross-sell opportunities, loyalty programs, or product recommendations. 
  • Withering Customers will have low recency but high monetary scores. These were once valuable customers that have since lapsed, try reinvigorating them with discounts, new product launches, or subscription options.

With segments like these available, marketing campaigns can be set up to target customers based on their RFM in an attempt to re-engage, retain, or refresh. ​​If you’re looking for predictive purchase analysis and need guidance on where to start, Relationship One is always here to help.